Rule engine and low code double-driven power grid diagram fusion management platform and method

Abstract

The application discloses a power grid number and graph fusion management platform and method driven by a rule engine and low code. The platform comprises an infrastructure layer, a data layer, an algorithm engine layer and a business application layer. The data layer is used for accessing and managing multi-source power grid data. The algorithm engine layer comprises a rule engine module configured to execute data verification, number and graph association and exception processing rules, a data quality diagnosis algorithm module for data quality diagnosis to identify abnormal data, and a number and graph fusion driving engine module for realizing two-way linkage between data and graphs. The business application layer comprises a low code arrangement platform for customizing management scene pages and work order processes, a number and graph integrated display module for displaying multi-source power grid data on the same screen, and a closed-loop work order management module for automatically generating and distributing management work orders. The application constructs a deep number and graph fusion system, improves abnormal positioning and management efficiency, realizes automatic control and whole-process tracing, does not need a large amount of code development, and shortens the response cycle of new business scenarios.

Description

Technical Field

[0001] This invention relates to a power grid data and graph fusion governance platform in the field of power systems, and more particularly to a power grid data and graph fusion governance platform driven by both a rule engine and low code, and also to a power grid data and graph fusion governance method driven by both a rule engine and low code. Background Technology

[0002] Power grid data governance is a core component ensuring the construction and stable operation of a digital twin power grid. Current mainstream power grid data governance solutions primarily rely on traditional data governance tools and graphical display systems. Specifically, data governance typically centers on database management systems, ETL (Extract, Transform, Load) tools, and data quality verification software. Data entry, cleaning, verification, and integration are achieved through form-based interfaces. Graphical display utilizes independent Geographic Information Systems (GIS) or dispatching map systems to present the spatial location, topological connections, and operational status of power grid equipment. This structured data and graphical data are stored in different systems, typically linked only loosely by key fields such as equipment IDs. The entire governance process relies heavily on manual processes or simple work order systems, lacking automated connections and global traceability between stages, from anomaly detection and notification for rectification to review and archiving.

[0003] However, the aforementioned existing technical solutions have significant drawbacks. First, the separate storage and display of data and graphics result in a shallow degree of "data-graphics fusion." Users need to frequently switch between data governance tools and graphics systems and manually link information, failing to obtain a unified view. This makes anomaly localization (such as topology connection errors) time-consuming and error-prone, leading to low efficiency in governance interaction. Second, the governance process fails to form an effective closed loop. From anomaly discovery to rectification and review, each stage relies on manual coordination. Rectification progress cannot be monitored in real time, results lack automatic verification, and the system fails to link and record changes in data and graphics before and after rectification, resulting in weak traceability throughout the process. Third, the system has poor business adaptability and high expansion costs. Adding new governance scenarios (such as access to new energy power plants) requires extensive customized code development for underlying verification rules, graphical interfaces, and work order logic, resulting in long response cycles and difficulty in meeting the rapidly changing needs of power grid operations. Finally, security control and governance processes are severely disconnected. Sensitive data is mostly anonymized using static methods, which cannot be flexibly adjusted according to roles and scenarios, easily leading to over-anonymization or insufficient protection. At the same time, audit logs fail to link governance details, making it difficult to meet compliance requirements for full lifecycle data traceability. Summary of the Invention

[0004] To address the technical problems of data-graph separation and poor adaptability in existing power grid data governance platforms and methods, this invention provides a power grid data-graph fusion governance platform and method driven by both a rule engine and low-code.

[0005] This invention is implemented using the following technical solution: a power grid data and graph fusion governance platform driven by both a rule engine and low-code, comprising: The infrastructure layer includes microservice clusters and containerized deployment modules, and provides hybrid distributed storage and caching; The data layer is used to access and manage multi-source power grid data, including power grid GIS topology maps; The algorithm engine layer includes: The rules engine module is used to configure and execute data validation, data graph association, and anomaly handling rules through a visual interface; The data quality diagnostic algorithm module is used to perform rule-based data quality diagnostics to identify data anomalies. The data-graphics fusion driving engine module is used to realize bidirectional linkage between data and graphics. The bidirectional linkage includes automatic graphic annotation of data anomalies and data response triggered by graphic operations. The business application layer includes: A low-code orchestration platform that allows users to customize governance scenario pages and work order processes by dragging and dropping components and templates; number Figure 1 An integrated display module is used to display the multi-source power grid data on the same screen. The closed-loop work order management module is used to automatically generate and allocate governance work orders based on data anomalies; The security desensitization module is used to execute dynamic desensitization strategies based on data sensitivity, user roles and operation scenarios, and / or use differential privacy technology to add noise to aggregated statistical data; The audit log and report module is used to collect and immutably store logs of user operations, system operations, and data changes. The logs include at least one of the following: operator, time, operation type, operation object, and operation content.

[0006] This invention establishes a two-way linkage logic between data and graphics through a deep data-graphics fusion mechanism. This enables automatic association of data anomalies with graphic annotations and triggering data responses through graphic operations. It breaks away from the traditional separation model of "data forms + independent graphics" and innovates into a scenario-based governance interaction of "managing data through graphics and verifying governance through graphics." Through a "dual-drive" architecture design, the rule engine supports visual configuration of data verification, data-graphics association, and anomaly handling rules, making the governance logic pluggable and enhancing adaptability. Meanwhile, the low-code platform provides power grid-specific components and scenario templates, eliminating the need for extensive code development and shortening the response cycle for new business scenarios. This solves the technical problems of data-graphics separation and poor adaptability in existing power grid data governance platforms and methods, achieving "visualized governance, scenario-based control, and secure traceability" of power grid data. It improves anomaly location and governance efficiency, realizes automated control and full traceability of the governance process, and ensures continuous improvement in data quality.

[0007] As a further improvement to the above solution, the microservice cluster includes multiple microservice modules that are deployed independently and communicate with each other through API or event bus; the containerized deployment module is used to containerize and orchestrate microservices, and to realize the visual monitoring and operation and maintenance of containers; the distributed storage and cache adopts a hybrid storage architecture of time-series database, relational database, graph database and distributed file system.

[0008] As a further improvement to the above scheme, the multi-source power grid data includes structured data, unstructured data, and spatiotemporal graphic data including the power grid GIS topology map.

[0009] As a further improvement to the above scheme, the data layer includes; A multi-source data access module is used to access the multi-source power grid data, realize data format conversion, data cleaning, and push the cleaned data to the corresponding storage system; the multi-source power grid data includes real-time streaming data, offline batch data, and heterogeneous interface data. The spatiotemporal data management module is used to integrate a GIS engine, manage the power grid GIS topology map and equipment spatial coordinate data, and provide spatial query and analysis functions. The spatiotemporal data management module supports the access and association of graphic data, including UAV inspection images and dispatch maps, and establishes a mapping relationship between spatiotemporal graphic data and equipment entities.

[0010] Furthermore, the rule engine module constructs a governance rule center based on the Drools rule engine and automatically generates DRL scripts after rule configuration; the data quality diagnosis algorithm module performs four-dimensional feature verification on the structured data and consistency verification on the data graph association by loading the data verification rules configured in the rule engine, automatically identifies data quality anomalies and marks the anomaly level; the data graph fusion driving engine module automatically locates the corresponding graph nodes, marks anomalies with colors / icons and pops up anomaly detail bubbles, and triggers data query, topology verification and batch verification when the user clicks / drags the graph nodes.

[0011] As a further improvement to the above solution, the anomaly levels include severe, general, and minor; severe anomalies are marked in red, general anomalies in yellow, and minor anomalies in orange; the data-graph fusion driving engine module pops up an anomaly details bubble next to the graph node, displaying the anomaly type, current value, standard value, and rectification suggestions.

[0012] As a further improvement to the above solution, the low-code orchestration platform provides a variety of power grid-specific components and scenario templates, and supports the customization of governance scenario pages and work order processes through drag-and-drop; the data Figure 1The integrated display module also supports flying line coloring, topology anomaly location, and one-click penetrating query functions; the closed-loop work order management module is built on a workflow engine and supports automatic creation, allocation, tracking, review, and archiving of work orders.

[0013] As a further improvement to the above solution, the power grid data map fusion governance platform also includes a user layer, which includes: A role-based user interface, which is used to provide an operation and maintenance interface for operation and maintenance personnel; The message notification module is used to push work order reminders, abnormal alarms, and report notifications via SMS, email, and system messages, and supports message read / unread marking and historical message query.

[0014] This invention also provides a power grid data graph fusion governance method driven by both a rule engine and low-code. This method applies any of the aforementioned power grid data graph fusion governance platforms driven by both a rule engine and low-code. The power grid data graph fusion governance method includes: Data access and preprocessing steps: Access multi-source power grid data, perform preprocessing, and store them according to type; Data fusion and quality diagnosis steps: The structured data and power grid graphic data are associated and fused, and data quality diagnosis is performed based on pre-configured rules. The diagnosed anomalies are marked on the graphical interface. Closed-loop work order processing steps: Automatically generate and assign governance work orders based on data anomalies, track the rectification process, and review and archive the rectification results; Security control and traceability steps: Throughout the entire governance process, sensitive data is dynamically anonymized and differentially protected for privacy, and a full-process audit log is recorded.

[0015] Compared to existing power grid data governance platforms and methods, the power grid data fusion governance platform and method driven by both rule engine and low-code in this invention have the following advantages: 1. This rule engine and low-code dual-driven power grid data-graph fusion governance platform establish a two-way linkage logic between data and graph through a deep data-graph fusion mechanism. This enables automatic association of data anomalies with graph annotations and triggering data responses through graph operations. It breaks the traditional separation mode of "data forms + independent graphs" and innovates into a scenario-based governance interaction of "managing data through graphs and verifying governance through graphs". Through the "dual-drive" architecture design, the rule engine supports visual configuration of data verification, data-graph association, and anomaly handling rules, making the governance logic pluggable and enhancing adaptability. The low-code platform provides power grid-specific components and scenario templates, eliminating the need for extensive code development and shortening the response cycle for new business scenarios. It solves the technical problems of data-graph separation and poor adaptability in existing power grid data governance platforms and methods, realizing "visualized governance, scenario-based control, and secure traceability" of power grid data. This improves anomaly location and governance efficiency, achieves automated control and full traceability of the governance process, and ensures continuous improvement in data quality.

[0016] 2. This rule engine and low-code-driven power grid data fusion governance platform, through the construction of a closed-loop management mechanism covering the entire process, and based on workflow engine and digital thread technology, seamlessly connects each stage of "anomaly detection - work order creation - responsibility allocation - rectification - review - archiving," and fully associates the entire chain of information "data - work order - graph - rule." This not only achieves automated control and seamless traceability of the governance process, making rectification progress visible in real time and rectification results automatically verified, but also accurately identifies governance responsibilities, thereby significantly improving the data rectification completion rate compared to traditional methods, fundamentally ensuring the continuous improvement and stable reliability of power grid data quality.

[0017] 3. This rule engine, coupled with a low-code-driven power grid data fusion governance platform, innovatively embeds security controls deep into the entire data governance lifecycle. Through an integrated security mechanism of "dynamic desensitization + differential privacy + full-process auditing," it effectively solves the persistent problem of security and governance being disconnected in traditional solutions. Dynamic desensitization flexibly adjusts strategies based on user roles and operational scenarios, avoiding the "one-size-fits-all" drawbacks of static desensitization; differential privacy technology injects noise during aggregation analysis, protecting individual privacy while ensuring the statistical analysis value of the data; and the fully tamper-proof audit logs meet compliance requirements, achieving traceability and auditability of governance operations.

[0018] 4. This rule engine and low-code-driven power grid data-graph fusion governance platform, in terms of system architecture and performance, adopts a hybrid distributed storage architecture based on microservices and containerization, possessing both high availability and elastic scalability. The collaborative work of time-series databases, graph databases, relational databases, and distributed file systems perfectly meets the storage and efficient querying needs of multi-source heterogeneous data in the power grid, enabling real-time governance operations for large-scale power grids with thousands of nodes or more. This solves the performance bottleneck of traditional monolithic architectures and single databases when facing massive spatiotemporal data and topological relationships.

[0019] 5. This rule engine, coupled with the low-code-driven power grid data fusion governance platform, forms a sustainable improvement loop of "governance-analysis-optimization." The platform can automatically generate multi-dimensional data governance effect reports and system operation reports, helping administrators accurately identify weak points in governance and thus driving continuous optimization of governance rules and scenario templates. This self-evolving capability based on data-driven decision-making enables the platform to continuously adapt to the dynamic development of power grid operations, building a long-term data quality management system with far-reaching application value.

[0020] 6. The rule engine and low-code dual-driven power grid data map fusion governance method have the same beneficial effects as the power grid data map fusion governance platform mentioned above, and will not be repeated here. Attached Figure Description

[0021] Figure 1 This is a system framework diagram of the power grid data graph fusion governance platform driven by both a rule engine and low code in Embodiment 1 of the present invention.

[0022] Figure 2 This is a flowchart of the power grid data graph fusion governance method driven by both rule engine and low code in Embodiment 2 of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0024] Current mainstream power grid data governance solutions mainly rely on traditional data governance tools, focusing on structured data form operations, supplemented by independent graphical display systems. The main technologies include the following.

[0025] 1. Data Governance Tools. These tools, centered around database management systems, ETL tools, and data quality verification software, enable data entry, cleaning, verification, and integration through a form-based interface. For example, they can filter out abnormal data using SQL statements, verify data quality based on preset rules (such as numerical range and field non-nullability), and generate quality reports in Excel format.

[0026] 2. Graphical Display System. An independent GIS platform and dispatching map system are used to display power grid graphical data. For example, the GIS system is used to present the spatial location and topological connections of equipment, while the dispatching map system is used to display the power grid's operational status. Graphical data and structured data are stored in different systems. The graphical system serves only as a visualization medium and does not participate in data governance logic.

[0027] 3. Data Association Method. A loose association between structured data and graphical data is established through key fields such as device ID. Users need to manually switch between the data governance tool and the graphical system for querying. After locating abnormal data in the data tool, the device ID needs to be recorded. Then, the ID needs to be entered into the graphical system to query the corresponding device graphical node to complete the "data-graphics" association verification.

[0028] 4. Inefficient process management. Governance processes rely on manual intervention or simple work order systems. After data anomalies are discovered, relevant personnel are notified to rectify them via email or paper documents. After rectification, the results are manually reported. There is a lack of automated control and traceability of the entire process of "anomaly discovery-rectification-review-archiving", and there is no linkage with graphical data to record changes in graphics before and after rectification.

[0029] 5. System scalability. Adding new business scenarios (such as data governance of new energy power plants and data governance of main grid OP interconnection) requires modifying the underlying code based on the traditional development model. For example, adding new data verification rules for new energy equipment requires the development of a new verification module. Customized development has a long cycle and it is difficult to respond quickly to changes in power grid business.

[0030] For example, in traditional power grid data governance, if a topology connection error is found on a 110kV line, maintenance personnel must first find the equipment ID of the line in the data quality report, then switch to the GIS system to input the ID to locate the line graphic, manually compare the graphic topology with the actual power grid connection, and notify the maintenance team to rectify the error via email. After rectification, the "data query-graphic verification" process must be repeated to verify the effect. The whole process is cumbersome, involves frequent switching between systems, and cannot track the rectification progress in real time.

[0031] The aforementioned power grid data governance technical solutions have four shortcomings: insufficient depth of data-map fusion, low governance efficiency, weak business adaptability, and disconnect between security and governance, as detailed below.

[0032] 1. Shallow integration of data and graphs leads to inefficient governance interaction. In existing solutions, data and graphs are stored and displayed separately, requiring users to manually link "data-graph" information across systems, making it impossible to intuitively obtain an integrated "graph-data" view. Data quality anomalies (such as topology errors or parameter mismatches) require form filtering and manual comparison of graphs to pinpoint the problem, which is time-consuming and error-prone, making it difficult to achieve an efficient "data governance by graph analysis" model.

[0033] 2. The governance process is not closed-loop, and traceability is weak. There is a lack of an automated, end-to-end closed-loop management mechanism. Each stage of data anomaly detection, rectification, and review relies on manual coordination. Rectification progress is not monitored in real time, and rectification results are not automatically verified. Furthermore, the "data-work order-graphics-rules" linkage has not been established, making it impossible to trace the governance process of abnormal data (such as changes in data values ​​before and after modification, and changes in related graphics), and making it difficult to pinpoint governance responsibility.

[0034] 3. Poor business adaptability and high expansion costs: Under the traditional development model, adding new governance scenarios (such as data access to new energy power plants and data integration of distribution network operation and dispatch) requires a lot of customized development, including adding data verification rule code, developing dedicated graphical display interfaces, and adjusting work order process logic. The average response cycle is more than 2 months, which cannot meet the needs of rapid changes in power grid business. Some low-code technology applications only cover the front-end page, and the back-end governance rules and processes still need to be developed with code, which limits scalability.

[0035] 4. Disconnect between security and governance, and insufficient compliance: Sensitive data (such as 500kV main network topology and user electricity consumption information) lacks dynamic protection throughout the entire governance process. Static desensitization is often used (such as fixed hiding of the last 4 digits of the ID card), which cannot flexibly adjust the desensitization strategy according to user roles and operation scenarios. This can easily lead to "over-desensitization affecting data availability and insufficient desensitization causing security risks." Audit logs only record basic operations (such as login time and operator) and do not associate them with data governance details, which cannot meet the "data lifecycle traceability" requirement of the Information Security Protection Standard 2.0.

[0036] Based on this, the present invention provides the following embodiments to solve the above problems, which can deeply integrate data graphs, realize closed-loop governance, have rapid expansion capabilities, and embed security mechanisms.

[0037] Example 1 Please see Figure 1 This embodiment provides a power grid data-graph fusion governance platform driven by both a rule engine and low-code. The platform mainly consists of a five-layer architecture: infrastructure layer, data layer, algorithm engine layer, business application layer, and user layer. Each layer collaborates to achieve data-graph fusion governance functions. Of course, in other embodiments, the platform may also include other logical architecture layers depending on actual needs.

[0038] The infrastructure layer includes a microservice cluster and a containerized deployment module, and provides hybrid distributed storage and caching to support elastic system scaling. In this embodiment, the microservice cluster includes multiple independently deployed microservice modules that communicate via APIs or an event bus. The containerized deployment module is used to containerize and orchestrate microservices, and to enable visual monitoring and maintenance of containers. The distributed storage and caching adopts a hybrid storage architecture of time-series databases, relational databases, graph databases, and distributed file systems.

[0039] Specifically, the multiple microservice modules can be split into 12 core microservice modules based on the Spring Cloud framework. Each module is deployed independently and communicates through REST API / event bus, including user authentication service, data access service, data graph fusion service, rule engine service, low-code orchestration service, work order management service, security de-identification service, audit log service, spatiotemporal data service, report statistics service, message notification service, and system management service.

[0040] The containerized deployment module can use Docker to package microservices into containers, and use DockerCompose to implement multi-container orchestration, define the container startup order (e.g., the spatiotemporal data service starts before the data-graph fusion service), network configuration, and data volume mounting; and use Portainer to implement container visualization monitoring and operation and maintenance, supporting CPU / memory usage viewing, log retrieval, and container restart.

[0041] Distributed storage and caching can adopt a hybrid storage architecture of "time-series database (InfluxDB) + relational database (MySQL) + graph database (Neo4j) + distributed file system (MinIO)"—InfluxDB stores real-time operating data such as current and voltage, MySQL stores structured data such as device parameters and work order information, Neo4j stores power grid topology data, and MinIO stores unstructured data such as device images and inspection videos; Redis is used to implement distributed caching, caching user permissions and frequently queried data to reduce database access pressure.

[0042] The data layer is used to access and manage multi-source power grid data, including power grid GIS topology maps. This multi-source power grid data includes structured data, unstructured data, and spatiotemporal graphical data containing power grid GIS topology maps. The data layer enables the mixed storage and management of multi-source power grid data (structured / unstructured / spatiotemporal data). In this embodiment, the data layer includes a multi-source data access module and a spatiotemporal data management module.

[0043] The multi-source data access module is used to access multi-source power grid data, perform data format conversion and data cleaning, and push the cleaned data to the corresponding storage system. Here, multi-source power grid data includes real-time streaming data (such as Kafka streams), offline batch data (such as CSV files), and heterogeneous interface data (such as GIS system APIs and EMS system interfaces). The multi-source data access module supports multi-source power grid data, performs data format conversion (such as JSON to XML conversion), preliminary cleaning (such as removing duplicate data and completing missing fields), and pushes the data to the corresponding storage system.

[0044] The spatiotemporal data management module integrates with the GIS engine to manage power grid GIS topology maps and equipment spatial coordinate data. It also provides spatial query functions (such as filtering equipment by region) and analytical functions (such as equipment density statistics and topology connectivity analysis). The spatiotemporal data management module can support the access and association of graphic data, including UAV inspection images and dispatch maps, and establish a mapping relationship between spatiotemporal graphic data and equipment entities (such as binding equipment IDs to GIS node IDs).

[0045] The algorithm engine layer includes a rules engine module, a data quality diagnostic algorithm module, and a data-graph fusion driving engine module, providing core capabilities for governance logic.

[0046] The rules engine module is used to configure and execute data validation, data-graph association, and anomaly handling rules through a visual interface. In this embodiment, the rules engine module can build a governance rule center based on the Drools rules engine and automatically generate DRL scripts after rule configuration. Data validation rules allow drag-and-drop selection of value / rule / relationship / attribute validation dimensions, configuring logic such as "current > 2000A" and "missing topology connection." Data-graph association rules can define association logic such as "matching device voltage level with graphic layer." Anomaly handling rules can set strategies such as "automatically correcting minor numerical anomalies to historical averages" and "generating rectification work orders for serious anomalies." These three types of rules automatically generate DRL scripts after configuration and load and take effect in real time.

[0047] The data quality diagnostic algorithm module is used to perform rule-based data quality diagnosis to identify data anomalies. It integrates a four-dimensional feature verification algorithm, supporting cross-system data correlation verification (such as consistency verification between dispatch EMS and distribution network GIS data) and real-time map-model comparison (such as difference identification between dispatch map models and GIS map models). It automatically identifies data quality anomalies and labels their anomaly levels. Specifically, the data quality diagnostic algorithm module can load data verification rules configured in the rule engine to perform four-dimensional feature verification (value features, rule features, relationship features, and attribute features) on structured data, perform consistency verification on data-map relationships, and automatically identify data quality anomalies and label their anomaly levels.

[0048] The data-graphics fusion engine module enables bidirectional interaction between data and graphics. This bidirectional interaction includes automatic graphic annotation of data anomalies and data response triggered by graphic operations, i.e., data change-graphic linkage and graphic operation-data response linkage. Specifically, the data-graphics fusion engine module can automatically locate corresponding graphic nodes, mark anomalies with colors / icons (e.g., red for excessive current, yellow for missing topology) and pop up an anomaly details bubble. When the user clicks / drags a graphic node, it triggers data queries (e.g., viewing associated device parameters), topology verification (e.g., updating GIS coordinates and verifying topology after adjusting position), and batch checks (e.g., selecting devices to verify voltage level consistency).

[0049] Traditional solutions require manual cross-system data-graphic association, and anomaly location relies on manual comparison. This embodiment, however, achieves bidirectional linkage through a data-graphic fusion engine—automatic annotation of graphic nodes for data anomalies and triggering data responses through graphic operations. Combined with a "one-click penetrating query" function, data from multiple systems can be retrieved simply by clicking on the graphic. In scenarios where operational and distribution data are integrated, the efficiency of locating "errors in associating routes with user profiles" is significantly improved compared to traditional manual comparison, completely revolutionizing the inefficient "form-based data retrieval" model.

[0050] In this way, a deep data-data fusion system is built to achieve integrated display and coordinated governance of "graphics and data", breaking the traditional separation mode of "data form + independent graphics" and transforming data governance from a form mode to a scenario-based interactive mode of "governing data by looking at graphics and verifying governance by graphics", thereby improving the efficiency of anomaly location and governance.

[0051] The anomaly levels are categorized as severe, moderate, and minor. Severe anomalies are marked in red, moderate anomalies in yellow, and minor anomalies in orange. The data-graphics fusion engine module displays an anomaly details bubble next to the graphics node, showing the anomaly type, current value, standard value, and rectification suggestions.

[0052] The business application layer includes low-code orchestration platforms, data... Figure 1 The integrated display module and closed-loop work order management module may also include a security masking module and an audit log and reporting module. The business application layer can build data through a low-code orchestration platform. Figure 1 Applications include integrated display, cross-system correlation verification, and closed-loop work order management.

[0053] The low-code orchestration platform is used to customize governance scenario pages and work order processes through drag-and-drop components and templates. Specifically, the platform provides a variety of power grid-specific components and scenario templates, and supports drag-and-drop customization of governance scenario pages and work order processes, solving the problems of slow response and weak scalability in traditional development models. In this embodiment, the low-code orchestration platform can be built on Vue + ElementPlus, providing 30+ power grid-specific components (such as topology diagram components, data table components, anomaly annotation components, work order components, and report components) and 10+ preset scenario templates (such as main grid equipment governance templates, distribution network operation and dispatch data integration templates, and new energy station access templates). Business personnel can customize governance scenario pages through drag-and-drop, configuring page layout, component attributes, and data association logic. The low-code orchestration platform supports the BPMN 2.0 standard process designer, allowing for visual configuration of work order processes (such as "anomaly detection → work order creation → responsibility assignment → rectification → review → archiving").

[0054] number Figure 1 The integrated display module is used to display multi-source power grid data on the same screen, supporting features such as flying line coloring, topology anomaly location, and one-click penetration query. Figure 1 The integrated display module enables multi-source data to be displayed on the same screen in one integrated manner, integrating GIS topology maps, equipment operation data, parameter data, and anomaly annotation information. It supports real-time coloring of flying lines (flying lines corresponding to abnormal data are highlighted in red), automatic location of topology anomalies (accidental connection edges are marked based on graph calculation), and multi-source map model split-screen comparison (differences between scheduling map model and GIS map model are displayed synchronously). It provides a "one-click penetrating query" function, which allows users to query multi-source data such as EMS, GIS, and marketing system by clicking on graphic nodes.

[0055] The closed-loop work order management module is used to automatically generate and allocate governance work orders based on data anomalies. This module can be built on a workflow engine (such as Activiti) and supports automatic work order creation (automatic generation based on data anomalies, manual initiation by users), allocation (based on "anomaly type - responsible department" mapping rules), tracking (recording data before and after modification, uploading supporting materials), review (calling data quality diagnostic algorithms for secondary verification), and archiving (generating rectification reports and storing them in the knowledge base). This design creates a full-process closed-loop management mechanism, combining digital thread technology to establish a "data-work order-graphics-rules" linkage, achieving automated control and full traceability of the governance process. The governance process is traceable, responsibility is identifiable, and continuous improvement in data quality is guaranteed.

[0056] The security masking module is used to execute dynamic masking strategies based on data sensitivity, user roles, and operational scenarios, and / or employ differential privacy technology to add noise to aggregated statistical data. The module can build a dynamic masking engine that identifies sensitive data through rule matching and machine learning, categorizing it into three levels based on sensitivity (Level 1: user privacy, mainnet topology; Level 2: core device parameters, power load; Level 3: non-core parameters, public information). Masking strategies are configured according to user roles and operational scenarios (e.g., hiding key nodes when ordinary maintenance personnel view Level 1 data, while administrators see the full data). The module can also use a Laplace mechanism to achieve differential privacy protection, adding minor noise to aggregated statistics (e.g., the number of abnormal devices in a region) to balance data security and availability.

[0057] The audit log and report module is used to collect and immutably store logs of user operations, system operations, and data changes. Logs include at least one of the following: operator, time, operation type, operation object, and operation content. The audit log and report module uses immutable storage and supports multi-dimensional log queries and security alerts (such as abnormal access to sensitive data or batch data modification). The audit log and report module can automatically generate data governance effectiveness reports (anomaly repair rate, data accuracy trend) and device status statistical reports, supporting export and visualization.

[0058] Traditional solutions employ static data masking, resulting in a disconnect between sensitive data protection and governance processes, and audit logs lacking detailed correlation. This embodiment addresses this by using dynamic data masking with differentiated protection based on roles / scenarios (e.g., hiding critical mainnet nodes for ordinary users), differential privacy to ensure aggregated data security, and full-process auditing recording of the "operation-data-graphics" correlation information, meeting the Level 3 requirements of the Cybersecurity Classified Protection 2.0 standard. This avoids the problems of "excessive masking affecting usability and insufficient masking causing risks," achieving a balance between data security and governance efficiency. Thus, through dynamic data masking (differentiated protection based on roles / scenarios), differential privacy (noise protection for aggregated data), and full-process auditing (unalterable logs), it achieves full lifecycle protection of sensitive data and traceable governance operations, covering the entire data governance lifecycle, balancing data availability and security, and meeting the requirements of the Cybersecurity Classified Protection 2.0 standard.

[0059] The user layer includes a role-based operation interface and a message notification module, which provides customized operation interfaces for users with different roles (operation and maintenance personnel, administrators, and schedulers).

[0060] The role-based operation interface is designed to provide operation and maintenance personnel with a user interface. It offers a graphical governance and work order processing interface for operations and maintenance personnel, supporting anomaly location and rectification. For administrators, it provides a rule configuration and system management interface, supporting rule maintenance and user permission management. Finally, for dispatchers, it provides a real-time monitoring and data query interface, supporting viewing operational data and cross-system data penetration queries.

[0061] The message notification module is used to push work order reminders, abnormal alarms, and report notifications via SMS, email, and system messages, and supports message read / unread marking and historical message query.

[0062] Traditional solutions require more than two months of customized development for adding new business scenarios. However, this embodiment uses a dual-drive architecture of "rule engine + low code" to support visual configuration of rules and drag-and-drop customization of scenarios. The newly added "new energy power station data governance" scenario has been tested and found to take only 10 days, shortening the cycle by 83%. The rules take effect in real time after configuration without restarting the system, which can quickly respond to changes in power grid business (such as the need to verify data of new energy storage devices) and solve the pain points of "difficult code modification and slow expansion" in traditional architectures.

[0063] Traditional solutions rely on manual coordination of the governance process, with no automatic verification of rectification results, resulting in a completion rate of only 75% or less. In contrast, this embodiment achieves full-process automation based on a workflow engine—automatic generation of work orders for anomalies, automatic allocation of work orders, and automatic review after rectification. Combined with digital thread technology to link the "data-work order-graphics" chain, it is possible to trace changes in data and graphics before and after rectification. The rectification completion rate has been tested and increased from 75% to 98.5%, ensuring continuous optimization of data quality.

[0064] In the current construction of digital twin power grids, "separation of data and maps, inefficient governance, poor adaptability, and disconnect from security" are common challenges faced by the industry. Companies like State Grid and China Southern Power Grid invest billions of yuan annually in data governance, but traditional solutions have limited effectiveness. The governance platform in this embodiment can be directly applied to core scenarios such as the construction of a unified power grid map, the integration of operation, distribution, and dispatch data, and the access of new energy power station data, indicating a clear and urgent market demand. Graphical governance reduces cross-system operations, improves anomaly location efficiency by 80%, and reduces manual verification workload by 50%. Taking a provincial power grid company as an example, it is estimated that it can save over 2 million yuan in operation and maintenance manpower costs annually.

[0065] In addition, it should be noted that the power grid data governance platform provided in this embodiment is not only applicable to the data governance of power transmission and transformation equipment, but can also be extended to scenarios such as distribution networks, new energy power plants (photovoltaic / wind power), and energy storage systems. By adding components and templates through the low-code platform, it can quickly adapt to future business needs such as "digital twin substations" and "intelligent power grid dispatching". At the same time, the core technologies (data-map fusion, dynamic desensitization) can be migrated to asset data governance in fields such as rail transit and municipal pipeline networks, and the market boundaries can be further expanded.

[0066] This power grid data governance platform integrates technologies from multiple fields, including microservices, low-code, graph computing, and data security, forming a technological barrier of "deep data-graph collaboration + dual-drive architecture + integrated security and governance." Compared to traditional data governance tools, this embodiment is more aligned with the characteristics of power grid operations (such as GIS topology association and power grid equipment parameter verification), making it difficult to replace with general-purpose tools. Furthermore, this embodiment provides high-quality data input for the digital twin power grid, ensuring the accuracy of power grid simulation and the reliability of real-time status perception. It can serve as a core support platform for the digital transformation of the power grid, helping power grid companies achieve the goals of "source-grid-load-storage" coordination and carbon neutrality.

[0067] It should be noted that if the new power grid data governance platform is designed using the following workarounds, there will also be compromises in terms of technical effectiveness or compatibility, as detailed in the following analysis.

[0068] (1) Data-Graph Integration Avoidance Solution. Instead of deep integration with "two-way linkage", this solution uses a superficial association of "passive data display in graphics"—only embedding a data query entry point in the graphics system. Users need to manually click on the graphics nodes to trigger the data query. There is no automatic annotation for data anomalies, and the graphics operation triggers the data response function. Although this solution avoids the protection point of "two-way linkage", the governance still relies on manual operation. The efficiency is not fundamentally different from the traditional solution, and it cannot achieve the core goal of "governing data by looking at graphics".

[0069] (2) Rule and Low-Code Architecture Bypass Solution. Instead of using the dual-drive architecture of "visual rule engine + low-code platform", a "code-based rule + fixed page" mode is adopted. New validation rules require writing Java / Python code, and new scenarios require developing dedicated pages; or only low-code front-end customization is retained, and back-end rules still require code development. Although this solution avoids the protection point of "dual drive", the business response cycle returns to more than 2 months in the traditional way, which cannot quickly adapt to changes in power grid business and has weak scalability.

[0070] (3) Closed-loop management mechanism avoidance scheme. Instead of adopting the "digital thread + automatic review" closed loop, a "manual work order + manual verification" mode is used. After the work order is created, it needs to be manually assigned. After the rectification is completed, the data and graphics need to be manually compared to confirm the effect. There is no automatic review and full-link traceability. Although this scheme avoids the protection point of "automated closed loop", the rectification completion rate drops to about 75%, which cannot guarantee the stability of data quality. Moreover, there is no detailed traceability of the governance process, making it difficult to locate responsibility.

[0071] (4) Security control technology avoidance scheme. Instead of "dynamic desensitization + differential privacy", a "static desensitization + basic audit" mode is adopted - sensitive data fields are fixedly hidden (such as the last 4 digits of ID card, the last 2 digits of core device parameters), and the audit log only records the operator and time, without data modification values ​​and graphic association information. Although this scheme avoids the protection point of "dynamic security control", it is prone to "excessive desensitization affecting data availability (such as hiding key parameters, making it impossible to judge the device status)" or "insufficient desensitization causing risks (such as the main network topology being fully displayed to ordinary users)", and it is impossible to trace the details of data modification, which does not meet the compliance requirements.

[0072] (5) Storage and Architecture Avoidance Solutions. Instead of adopting the "hybrid storage + microservices" architecture, a "single relational database + monolithic application" model is used—all data (real-time / structured / graph) is stored in MySQL, and the system is deployed as a monolith; or instead of using a graph database to store topology relationships, relational database tables are used to simulate the topology. Although this solution avoids the protection point of "hybrid storage + microservices", the real-time data write performance is insufficient (InfluxDB writes tens of thousands per second vs MySQL writes thousands per second), the topology query efficiency is low (Neo4j graph query vs relational database multi-table joins), the system cannot support 1000+ users online at the same time, and it is not suitable for the needs of large-scale power grid governance.

[0073] In summary, designing a power grid data governance platform based on the above five aspects cannot achieve the technical effects obtained in this embodiment. Therefore, compared with existing power grid data governance platforms, the power grid data fusion governance platform driven by both a rule engine and low-code has the following significant advantages: 1. This rule engine and low-code dual-driven power grid data-graph fusion governance platform establish a two-way linkage logic between data and graph through a deep data-graph fusion mechanism. This enables automatic association of data anomalies with graph annotations and triggering data responses through graph operations. It breaks the traditional separation mode of "data forms + independent graphs" and innovates into a scenario-based governance interaction of "managing data through graphs and verifying governance through graphs". Through the "dual-drive" architecture design, the rule engine supports visual configuration of data verification, data-graph association, and anomaly handling rules, making the governance logic pluggable and enhancing adaptability. The low-code platform provides power grid-specific components and scenario templates, eliminating the need for extensive code development and shortening the response cycle for new business scenarios. It solves the technical problems of data-graph separation and poor adaptability in existing power grid data governance platforms and methods, realizing "visualized governance, scenario-based control, and secure traceability" of power grid data. This improves anomaly location and governance efficiency, achieves automated control and full traceability of the governance process, and ensures continuous improvement in data quality.

[0074] 2. This rule engine and low-code-driven power grid data fusion governance platform, through the construction of a closed-loop management mechanism covering the entire process, and based on workflow engine and digital thread technology, seamlessly connects each stage of "anomaly detection - work order creation - responsibility allocation - rectification - review - archiving," and fully associates the entire chain of information "data - work order - graph - rule." This not only achieves automated control and seamless traceability of the governance process, making rectification progress visible in real time and rectification results automatically verified, but also accurately identifies governance responsibilities, thereby significantly improving the data rectification completion rate compared to traditional methods, fundamentally ensuring the continuous improvement and stable reliability of power grid data quality.

[0075] 3. This rule engine, coupled with a low-code-driven power grid data fusion governance platform, innovatively embeds security controls deep into the entire data governance lifecycle. Through an integrated security mechanism of "dynamic desensitization + differential privacy + full-process auditing," it effectively solves the persistent problem of security and governance being disconnected in traditional solutions. Dynamic desensitization flexibly adjusts strategies based on user roles and operational scenarios, avoiding the "one-size-fits-all" drawbacks of static desensitization. Differential privacy technology injects noise during aggregation analysis, protecting individual privacy while ensuring the statistical analysis value of the data. The fully tamper-proof audit logs meet compliance requirements, achieving traceability and auditability of governance operations.

[0076] 4. This rule engine and low-code-driven power grid data-graph fusion governance platform, in terms of system architecture and performance, adopts a hybrid distributed storage architecture based on microservices and containerization, possessing both high availability and elastic scalability. The collaborative work of time-series databases, graph databases, relational databases, and distributed file systems perfectly meets the storage and efficient querying needs of multi-source heterogeneous data in the power grid, enabling real-time governance operations for large-scale power grids with thousands of nodes or more. This solves the performance bottleneck of traditional monolithic architectures and single databases when facing massive spatiotemporal data and topological relationships.

[0077] 5. This rule engine, coupled with the low-code-driven power grid data fusion governance platform, forms a sustainable improvement loop of "governance-analysis-optimization." The platform can automatically generate multi-dimensional data governance effect reports and system operation reports, helping administrators accurately identify weak points in governance and thus driving continuous optimization of governance rules and scenario templates. This self-evolving capability based on data-driven decision-making enables the platform to continuously adapt to the dynamic development of power grid operations, building a long-term data quality management system with far-reaching application value.

[0078] Example 2 Please see Figure 2 This embodiment provides a power grid data graph fusion governance method driven by both a rule engine and low-code. This power grid data graph fusion governance method applies the power grid data graph fusion governance platform driven by both a rule engine and low-code in Embodiment 1. The power grid data graph fusion governance method includes the following steps.

[0079] Data access and preprocessing steps: Access multi-source power grid data, perform preprocessing, and store them according to type.

[0080] Data fusion and quality diagnosis steps: The structured data and power grid graphic data are associated and fused, and data quality diagnosis is performed based on pre-configured rules. The diagnosed anomalies are marked on the graphical interface.

[0081] Closed-loop work order processing steps: Automatically generate and assign governance work orders based on data anomalies, track the rectification process, and review and archive the rectification results.

[0082] Security control and traceability steps: Throughout the entire governance process, sensitive data is dynamically anonymized and differentially protected for privacy, and a full-process audit log is recorded.

[0083] Example 3 This embodiment provides a power grid data graph fusion governance method driven by both a rule engine and low-code, which applies the power grid data graph fusion governance platform driven by both a rule engine and low-code in Embodiment 1. The power grid data graph fusion governance method includes the following steps.

[0084] Data access and preprocessing steps include: (1) Connect to the data sources of the power grid EMS system, GIS system and marketing system to receive real-time operation data, equipment parameter data, graphic data (GIS topology map, equipment image) and offline batch data; (2) Perform format conversion (e.g., convert XML format data output by the GIS system to JSON format) and data cleaning (remove duplicate data and fill in non-critical missing fields with default values) on the received data to obtain preprocessed data; (3) Push the preprocessed data to the corresponding storage system according to the type; among them, real-time data is written to the time series database: InfluxDB, structured data is written to the relational database: MySQL, topological relational data is written to the graph database: Neo4j, unstructured data is written to the distributed file system: MinIO, and graph data is associated with the device ID and then stored in the spatiotemporal data management module of the data layer.

[0085] The steps for data-image fusion and quality diagnosis include: (1) The GIS topology map of the power grid is loaded by calling the GIS engine through the spatiotemporal data management module. The data fusion service associates the equipment parameter data in MySQL and the real-time running data in InfluxDB with the graphic node according to the mapping relationship of "equipment ID-graphic node ID" to realize the integrated display of "graphics-data". (2) Load the data verification rules configured in the rule engine through the data quality diagnosis algorithm, perform four-dimensional feature verification on the structured data (such as verifying whether the voltage value is within a reasonable range and whether the equipment topology connection is complete), and perform consistency verification on the data-graph association relationship (such as whether the equipment voltage level matches the graphic layer). (3) Receive quality diagnosis results through the data fusion driving engine module. If data abnormality is found, the corresponding graphic node is automatically located. Serious abnormalities are marked in red, general abnormalities are marked in yellow, and minor abnormalities are marked in orange. At the same time, an abnormality details bubble pops up next to the graphic node, showing "abnormality type, current value, standard value, and rectification suggestions". (4) Synchronize abnormal data information to the work order management module and trigger the work order creation process: serious abnormalities generate work orders immediately, and general / minor abnormalities generate work orders in batches every hour. The work order automatically carries "abnormal data ID, graphic location information, and abnormal description".

[0086] The rule configuration and scenario customization steps (this step is optional and should be performed as needed) include: (1) If you need to add a new governance scenario (such as data governance of new energy power stations), select “New Energy Power Station Access Template” through the low-code orchestration platform, and drag and drop the topology diagram component, data table component, and work order component to adjust the page layout; (2) Configure new energy equipment-specific verification rules (such as photovoltaic inverter power range, wind power equipment speed threshold) and data-graph association rules (such as photovoltaic array graph nodes and inverter data binding) in the rule engine center. (3) Adjust the work order process through the BPMN process designer and set relevant parameters such as "allocate new energy abnormal work orders to new energy operation and maintenance teams" and "rectification time limit 48 hours"; (4) Generate scenario configuration scripts through the low-code orchestration platform and synchronize them to the business application layer; new scenarios can take effect without restarting the system, and users can carry out governance work by selecting the corresponding scenario in the role-based operation interface.

[0087] The closed-loop work order processing steps specifically include: (1) The closed-loop work order management module automatically assigns work orders to the corresponding departments (e.g., topology anomalies are assigned to the operation and maintenance department, and parameter anomalies are assigned to the dispatch center) based on the mapping rule of “anomaly type-responsible department”, and supports the administrator to manually adjust the assignment results; (2) Send SMS / system messages to the person in charge of the work order through the message notification module at the user level, informing them of the "work order number, exception description, and processing time limit", and at the same time, send SMS / system messages to the person in charge of the work order through the message notification module at the user level. Figure 1 The integrated display interface indicates the status of the work order (such as "Pending" or "Processing"). (3) The person in charge logs into the operation and maintenance personnel interface, clicks on the abnormal node in the graph, and uses "one-click penetrating query" to view the associated multi-source data (such as historical operation data in the scheduling EMS and user profile data in the marketing system), and uses the rectification tools provided by the platform (data editing interface, batch correction template) to complete the modification of abnormal data; (4) Automatically record “data value before modification, data value after modification, modification time, and modifier” and supporting materials (such as on-site photos, test reports, etc.) through the audit log and report module of the business application layer; after rectification, submit a review application; (5) The data quality diagnosis algorithm is called through the closed-loop work order management module to perform secondary verification on the rectification data; if the verification passes, the work order enters the archiving node; if it fails, the work order is returned to the rectification node with review comments (such as "the voltage level is still not compatible with the line"); the review personnel can view the rectification records and graphic changes and manually confirm the rectification effect. (6) After the work order is archived, a rectification report is automatically generated, which includes "basic information of the work order, rectification effect (abnormal repair rate, data accuracy improvement), and related data changes". The report is stored in the work order knowledge base and supports query and statistics by abnormal type and processing department.

[0088] Safety control and traceability steps specifically include: (1) When a user accesses the system, the user authentication service completes identity verification based on the OAuth2.0 protocol and assigns operation permissions through the RBAC (role-based access control) model (e.g., maintenance personnel can only view sensitive data at level 2 and below, while administrators can access all data); when a user triggers a sensitive data access operation (e.g., viewing the 500kV main network topology or user electricity information), the security desensitization module of the business application layer automatically identifies the user role and operation scenario and calls the corresponding desensitization strategy to perform processing; for example, when ordinary maintenance personnel view a user's ID number, the middle 8 digits are hidden (format is 110101********1234), and key nodes are hidden when viewing the main network topology; when the administrator reviews the data, they can obtain the complete sensitive data; (2) When a user initiates an aggregation analysis request (such as statistics on “number of abnormal 110kV transformers in the western part of the city” or “monthly data rectification completion rate”), the noise value is calculated based on the preset privacy budget ε = 0.1-0.5 (the smaller the ε value, the stronger the privacy protection). Small noise is added to the aggregation result (such as returning 24 or 26 when the actual number of abnormalities is 25). At the same time, an independent privacy budget pool is maintained for each user. Each aggregation query consumes the corresponding budget (such as 0.1 budget for regional abnormality statistics). An application for replenishment must be submitted after the budget is exhausted. When the same query is repeated within 1 hour, the initial calculation result is reused to avoid repeated consumption of the budget. (3) Three types of logs are collected in real time through the audit log and report module of the business application layer: user operation logs (login / logout, rule configuration, work order creation / processing, data query / modification), system operation logs (data access, de-identification execution, rule loading, service start / stop), and data change logs (values ​​before / after abnormal data modification, modifier, modification time, associated device ID); the log fields include "operator, operation time, operation type, operation object, operation content, IP address, device information, operation result"; all logs are stored in a read-only database with timestamps to prevent tampering; (4) Security alarm rules are preset through the audit log and report module of the business application layer (such as "login failure ≥5 times within 10 minutes of the same IP address", "ordinary users attempt to access first-level sensitive data ≥3 times within 1 hour", "modification of ≥10 core parameters of the device in a single instance"). After the rules are triggered, the administrator is notified immediately via SMS and email, and a pop-up prompt is displayed on the system management interface. The logs are automatically analyzed weekly to identify potential risk operations and generate a "Security Analysis Report" which includes details of risk operations, users involved, and suggested optimization strategies.

[0089] The steps for performance statistics and continuous optimization include: (1) Periodically (e.g., daily / weekly / monthly) extract data from each storage system, obtain "number of abnormal work orders, rectification completion rate, and average processing time" from the work order management module, obtain "data accuracy and distribution of abnormal data types" from the data quality diagnosis module, and obtain "user operation frequency and sensitive data access records" from the audit log module; (2) Process the extracted data into visualization charts (line chart to show the data accuracy trend, bar chart to show the proportion of abnormal types, pie chart to show the distribution of work order processing departments), generate "Data Governance Effect Report" and "System Operation Report", and support export to Excel and PDF formats; (3) View reports through the system management interface, analyze weak links in governance (such as "high proportion of topology abnormal work orders" and "low data accuracy in distribution network areas"), and optimize the verification rules in the rule engine (such as adding secondary verification logic for topology connectivity) and adjust the low-code scenario template (such as optimizing the layout of distribution network data governance page components) to form a continuous improvement closed loop of "governance-analysis-optimization".

[0090] Example 4 This embodiment provides a computer terminal, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the power grid data-graph fusion governance method driven by both a rule engine and low-code in Embodiment 2 or 3.

[0091] When applying the methods of Embodiment 2 or 3, they can be implemented in software form, such as by designing them as stand-alone programs and installing them on a computer terminal, which can be a computer, smartphone, control system, or other IoT devices. Alternatively, the methods of Embodiment 2 or 3 can be designed as embedded programs and installed on a computer terminal, such as on a microcontroller.

[0092] Example 5 This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, it implements the steps of the power grid data-graph fusion governance method driven by both a rule engine and low-code in Embodiment 2 or 3.

[0093] When applying the method of Embodiment 2 or 3, it can be applied in the form of software, such as by designing it as a program that can run independently on a computer-readable storage medium. The computer-readable storage medium can be a USB flash drive, designed as a USB security token, and the program is designed to start the entire method through an external trigger.

[0094] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A power grid data-graph fusion governance system driven by both a rule engine and low-code, characterized in that, It includes: The infrastructure layer includes microservice clusters and containerized deployment modules, and provides hybrid distributed storage and caching; The data layer is used to access and manage multi-source power grid data, including power grid GIS topology maps; The algorithm engine layer includes: The rules engine module is used to configure and execute data validation, data graph association, and anomaly handling rules through a visual interface. The data quality diagnostic algorithm module is used to perform rule-based data quality diagnostics to identify data anomalies. The data-graphics fusion driving engine module is used to realize bidirectional linkage between data and graphics. The bidirectional linkage includes automatic graphic annotation of data anomalies and data response triggered by graphic operations. The business application layer includes: A low-code orchestration platform that allows users to customize governance scenario pages and work order processes by dragging and dropping components and templates; The integrated data and graph display module is used to display the multi-source power grid data on the same screen. The closed-loop work order management module is used to automatically generate and allocate governance work orders based on data anomalies; The security desensitization module is used to execute dynamic desensitization strategies based on data sensitivity, user roles and operation scenarios, and / or use differential privacy technology to add noise to aggregated statistical data; The audit log and report module is used to collect and immutably store logs of user operations, system operations, and data changes. The logs include at least one of the following: operator, time, operation type, operation object, and operation content.

2. The power grid data and graph fusion governance system driven by both rule engine and low-code as described in claim 1, characterized in that, The microservice cluster includes multiple independently deployed microservice modules that communicate with each other via API or event bus; the containerized deployment module is used to containerize and orchestrate microservices, and to realize the visual monitoring and operation and maintenance of containers; the distributed storage and caching adopts a hybrid storage architecture of time-series database, relational database, graph database and distributed file system.

3. The power grid data and graph fusion governance system driven by both rule engine and low-code as described in claim 1, characterized in that, The multi-source power grid data includes structured data, unstructured data, and spatiotemporal graphic data including the power grid GIS topology map.

4. The power grid data and graph fusion governance system driven by both rule engine and low-code as described in claim 1, characterized in that, The data layer includes; A multi-source data access module is used to access the multi-source power grid data, realize data format conversion, data cleaning, and push the cleaned data to the corresponding storage system; the multi-source power grid data includes real-time streaming data, offline batch data, and heterogeneous interface data. The spatiotemporal data management module is used to integrate a GIS engine, manage the power grid GIS topology map and equipment spatial coordinate data, and provide spatial query and analysis functions. The spatiotemporal data management module supports the access and association of graphic data, including UAV inspection images and dispatch maps, and establishes a mapping relationship between spatiotemporal graphic data and equipment entities.

5. The power grid data and graph fusion governance system driven by both rule engine and low-code as described in claim 3, characterized in that, The rule engine module builds a governance rule center based on the Drools rule engine and automatically generates DRL scripts after rule configuration; the data quality diagnosis algorithm module performs four-dimensional feature verification on the structured data and consistency verification on the data graph relationship by loading the data verification rules configured in the rule engine, and automatically identifies data quality anomalies and marks the anomaly level. The data-graph fusion driving engine module automatically locates the corresponding graph nodes, marks anomalies with colors / icons and pops up an anomaly details bubble, and triggers data query, topology verification and batch verification when the user clicks / drags the graph node.

6. The power grid data and graph fusion governance system driven by both rule engine and low-code as described in claim 5, characterized in that, The anomaly levels include severe, general, and minor; severe anomalies are marked in red, general anomalies in yellow, and minor anomalies in orange; the data-image fusion driving engine module pops up a bubble of anomaly details next to the image node, displaying the anomaly type, current value, standard value, and rectification suggestions.

7. The power grid data and graph fusion governance system driven by both rule engine and low-code as described in claim 1, characterized in that, The low-code orchestration platform provides a variety of power grid-specific components and scenario templates, and supports the customization of governance scenario pages and work order processes through drag-and-drop. The integrated data and graph display module also supports flying line coloring, topology anomaly location, and one-click penetrating query functions. The closed-loop work order management module is built on a workflow engine and supports the automatic creation, allocation, tracking, review, and archiving of work orders.

8. The power grid data and graph fusion governance system driven by both rule engine and low-code as described in claim 1, characterized in that, The power grid data map fusion governance platform also includes a user layer, which includes: A role-based user interface, which is used to provide an operation and maintenance interface for operation and maintenance personnel; The message notification module is used to push work order reminders, abnormal alarms, and report notifications via SMS, email, and system messages, and supports message read / unread marking and historical message query.

9. A power grid data-graph fusion governance method driven by both a rule engine and low-code, characterized in that, Its application is the power grid data graph fusion governance system driven by a rule engine and low code as described in any one of claims 1-8, wherein the power grid data graph fusion governance method includes: Data access and preprocessing steps: Access multi-source power grid data, perform preprocessing, and store them according to type; Data fusion and quality diagnosis steps: The structured data and power grid graphic data are associated and fused, and data quality diagnosis is performed based on pre-configured rules. The diagnosed anomalies are marked on the graphical interface. Closed-loop work order processing steps: Automatically generate and assign governance work orders based on data anomalies, track the rectification process, and review and archive the rectification results; Security control and traceability steps: Throughout the entire governance process, sensitive data is dynamically anonymized and differentially protected for privacy, and a full-process audit log is recorded.

Images